Cohesity CEO: We're more than a backup company

Enterprise storage company Cohesity is moving from backup to data management and from selling integrated appliances to a software-only business.

Cohesity CEO Mohit Aron said backup is just one use case of secondary storage, and it was a good starting point for the company. Now he wants to focus on uniting analytics, test/dev and all the other uses of secondary data on the Cohesity platform. His goal is "smartphone-like simplicity" for data management.

"We started with backups, but we are not just a backup company," Aron said. "We have an acronym for it -- JAB, just a backup. We are doing something much bigger. We're all about data management."

Aron also said the Cohesity DataPlatform software is now available without underlying hardware. Cohesity, based in San Jose, Calif., now only recognizes revenue from software, relying on partners to bundle and distribute it with hardware. That is a similar business model adopted by hyper-converged vendor Nutanix, another company that Aron helped start. Cohesity software runs on servers from Hewlett Packard Enterprise, Dell EMC and Cisco, and others.

Cohesity -- along with Veeam, Rubrik, Actifio and Druva -- is among the heavily funded private companies challenging established backup leaders Veritas, Dell EMC, IBM and Commvault. The industry veterans have expanded their products to add integrated scale-out data management platforms to match the relative newcomers.

We spoke with Aron about some of Cohesity's big changes over the past year or so, including a $250 million funding round and its acquisition of Imanis Data.

The Cohesity CEO hinted that more funding and more acquisitions may follow as the vendor tries to keep up with an expanding data protection and data management market.

Do you intend to go beyond analytics and test/dev -- the things that are considered secondary storage? You already have a built-in file system, will you get into NAS primary storage?

Mohit Aron: I like to envision the data center as an iceberg, where the top, visible part is the latency-sensitive stuff, where primary storage is. But 80% of the data center is the lower part of the iceberg, where backup, analytics and all the other secondary storage lives.

Mohit Aron

Technically, we can go into the primary space. But from a business perspective, that lower part of the iceberg hasn't seen any innovation for the last 10, 15 years, so we intend to stay there.

From a business point of view, these are two very different sales motions. In the top of the iceberg, you're selling IOPS. In the lower iceberg, you're selling insights from the data. There are some good companies working at the top of the iceberg, like my former company Nutanix and Pure Storage. We just partner with them.

We have decided to stay in that lower part of the iceberg and not go toward the upper part where these other companies operate. And we're doing phenomenally well, so we'll just keep doing that. So it's not in the near future, but who knows what's in the distant future. Never say never.

But Nutanix and Pure are also getting into the lower part of the iceberg with their own data protection products. What does that do to your strategy?

Aron: This segment of the iceberg requires focus, and I don't believe those companies will be able to refocus on it. Companies can dabble. Even Amazon can do snapshots and whatnot, but their focus is not on that. In fact, [Microsoft] Azure and Amazon encourage us to build on their platforms because they know that we have the focus.

[Nutanix and Pure] can dabble [in data protection], but if they do that too much, they'll run into us, and we have six years' worth of a lead on them in this space. Also, they lose focus on where they shine. ... So we're really not concerned.

It's the same thing with Nutanix and Pure. They can dabble, but if they do that too much, they'll run into us, and we have six years' worth of a lead on them in this space. Also, they lose focus on where they shine, which is the top of the iceberg. So we're really not concerned.

You have $410 million venture funding, you are valuated at more than $1 billion, and you say your business is good. Are you actively going after an IPO yet?

Aron: The future is not something we comment or fixate on.

Cohesity is still in growth right now, but when will you start to focus on profitability?

Aron: I believe one always has to keep in mind the economics and what makes for a profitable company. If you look at Amazon or Salesforce, they have a sustainable, mature business model that makes money for them. They keep investing in growth because as long as the product is good, whatever you invest in growth will become profit.

I think the business here at Cohesity is very solid. The mature part is returning money back to the company, and so we're just investing in growth. I would like to keep doing that for the foreseeable future. And if we ever need to dial it down and become profitable, we can. We can become profitable in about a year or so once we stop investing in growth.

Are you finished with venture capital, or do you expect to go back for more?

Aron: Since we are investing in growth, we probably will need more capital. We don't care if it comes from private markets or public markets. It all depends on what the future holds.

As you move from just backup to data management, who do you see as your competitors?

Aron: When we were just backup, we competed with anyone who is in backups. So we competed with Veritas, [Dell EMC] Avamar, [Dell] EMC NetWorker, Commvault and Veeam. But we also happened to compete with [Dell EMC] Data Domain, because if someone buys us, they really don't need a Data Domain appliance. But that doesn't mean we're an appliance company at all. We displaced those backup software vendors and predominantly our expertise is with software.

Any software has to run on hardware, so by virtue of that, we are on some hardware, but we are no longer selling that hardware ourselves. We offload all of the hardware to other vendors and distributors like Cisco, HPE and Dell. We are a software-centric company.

Cohesity was sold with your appliance or on hardware from a partner. Can I now buy just the software and use it on any hardware or disk target?

Aron: Any hardware that we have blessed. Like VMware, we have a hardware compatibility list, too. We don't want to be in the business of the customer buying any random hardware, because then we're just solving hardware bugs. And we have multiple SKUs, so customers are pretty happy with the choices they have.

But you do still make the appliance available?

Aron: Yes, but it's sold only through distributors now. It's not going to ship from Cohesity. We only ship software. So a channel partner will be the one who puts it together and sends it to a customer.

We started that motion earlier this year. There was a time when we shipped both hardware and software, but as we head into the new fiscal year, it's 100% all software now.

You've recently acquired Imanis Data. Are other acquisitions planned?

Aron: We will definitely evaluate technology as we come across it. Our vision is broad. There is so much we can do in the marketplace, either by partnering with companies or acquiring them. We keep our eyes open on what's a great synergy for us, but we don't like to comment on what we're looking at or what we might do in the future.

Do you see any product or service gaps Cohesity needs to fill?

Aron: We have done a pretty good job of 95% of what a traditional legacy backup product does. That 5% of what we sometimes might not be able to do, like an arcane production environment we don't support backups for today, we say they can continue using their legacy backup software, which can then write to our file system.

Enterprises can still use their legacy systems. We're not something that necessarily replaces all their environments. If they have some legacy storage, they can back it up using their legacy backup software, which can write to us. But we speed up 95% of that functionality onto our backup software. We can coexist with existing backup software like NetBackup or Commvault.

But there was a time when we did not support Solaris, and so our customers were using third-party backup software to back up Solaris and then write to us. Over time, we added support for backing up Solaris natively, so they didn't have to use their backup software. That's a great example of a gap that used to exist but now doesn't. But the nice thing is there are workarounds for those gaps.

We used to have a gap in backing up NoSQL stores, and now that gap is filled with the acquisition of Imanis Data. It's an old industry, there can always be gaps, but I think 95% of the customers now find it adequate or have no gaps at all.